Hiroyuki Hisamatsu

Fluid-based analysis of a network with DCCP connections and RED routers

In this paper, we model DCCP congestion control mechanism and RED as independent discrete-time systems using fluid-flow approximation. By interconnecting DCCP connections and RED routers, we model the entire network as a feedback system called DCCP/RED. We then analyze the steady state performance and the transient state performance of DCCP/RED. Specifically, we derive the packet transmission rate of DCCP connections, the packet transmission rate, the packet loss probability, and the average queue length of the RED router in steady state. Moreover, we investigate the parameter region where DCCP/RED operates stably by linearizing DCCP/RED around its equilibrium point. We also evaluate the transient state performance of DCCP/RED in terms of ramp-up time, overshoot, and settling time. Consequently, we show that the stability and the transient state performance of DCCP/RED degrade when the weight of the exponential weighted moving average, which is one of RED control parameters, is small. To solve this problem, by adding changes to the function with which RED determines the packet loss probability, we propose RED-IQI (RED with immediate queue information). We analyze the transient state performance of the feedback system DCCP/RED-IQI where DCCP connections and RED-IQI routers are interconnected. Consequently, we show that DCCP/RED-IQI has significantly better transient state performance than DCCP/RED

Design and Evaluation of Hybrid Congestion Control Mechanism for Video Streaming

In recent years, video streaming services over TCP, such as YouTube, have become more and more popular. TCP New Reno, the current TCP standard, performs greedy congestion control, which increases the congestion window size until packet loss occurs. Therefore, because TCP transmits data at a much higher rate than the video playback rate, the probability of packet loss in the network increases, which in turn takes bandwidth from other network traffic. In this paper, we propose a new transport-layer protocol, called TCP Stream, which solves the problem of TCP in video streaming. TCP Stream performs a hybrid congestion control that combines the loss-based congestion control, which uses packet loss as an index of congestion, and the delay-based congestion control, which uses delay as an index of congestion. Simulation and experimental results show that TCP Stream transmits data at the adjusted rate, unlike TCP New Reno, and does not steal bandwidth from other network traffic.

Steady state and transient state behaviours analyses of TCP connections considering interactions between TCP connections and network

The Internet uses a window-based congestion control mechanism in transmission control protocol (TCP). In the literature, there have been a great number of analytical studies on TCP. Most of those studies have focused on the statistical behaviour of TCP by assuming a constant packet loss probability in the network. However, the packet loss probability, in reality, changes according to the packet transmission rates from TCP connections. Conversely, the window size of a TCP connection is dependent on the packet loss probability in the network. In this paper, we explicitly model the interaction between the congestion control mechanism of TCP and the network as a feedback system. By using this model, we analyse the steady state and the transient state behaviours of TCP. We derive the throughput and the packet loss probability of TCP, and the number of packets queued in the bottleneck router. We then analyse the transient state behaviour using a control theoretic approach, showing the influence of the number of TCP connections and the propagation delay on the transient state behaviour of TCP. Copyright

Proposal and Evaluation of an Information Dissemination Method Based on Flooding for Energy Efficiency in Wireless Sensor Networks

energy efficient. In this paper, we propose a new flooding method-based information dissemination method that tackles the issue of energy-efficient information dissemination in WSNs. By interpreting same message broadcasts from an adjacent node to a sending node as an acknowledgment of successful message transmission, we control message broadcasts and decrease the number of messages transmitted and received. We show by means of simulations that, in contrast to the conventional flooding method, our proposed method results in an improvement in the information delivery ratio with only a marginal increase in power consumption when packet loss probability is high. We also show that our proposed method results in less energy consumption in the overall network than the flooding method when the wireless radio range is long.

Method for Countering Social Bookmarking Pollution Using User Similarities

In this paper, we propose a method for countering social bookmark pollution. First, we investigate the characteristics of social bookmark pollution and show that high similarities in the user bookmarks result in social bookmark pollution. Then, we discuss a bookmark number reduction method based on user similarities between the user bookmarks. We evaluate the proposed method by applying it to Hatena Bookmark. It is found that the proposed method only slightly reduces the bookmark number of the Web pages that are not affected by social bookmark pollution but greatly reduces the bookmark number of those Web pages that are affected by social bookmark pollution.

Evaluation of a Bicycle-Mounted Ultrasonic Distance Sensor for Monitoring Road Surface Condition

In the case of cycling at a blind road, at night or cycling by elderly people, it is helpful to prevent accidents if information on bad road surface condition, such as break, potholes, obstacles, bumps, is preliminary obtained. In this paper, we consider to use a low-cost ultrasonic distance sensor attached on a bicycle for monitoring the road surface conditions in the front area. We design and implement a road surface condition monitoring system using an off-the-shelf node Arduino, an ultrasonic distance sensor, and a typical bicycle. In the implemented system, to limit the sensing region, the ultrasonic distance sensor is covered by a plastic shield plate. Through experimental evaluations, we show that the monitoring system can detect the 223 cm away obstacle on road in the front area of moving bicycle.

Energy-efficient information dissemination based on received signal strength in wireless sensor networks

Power-saving is one of the important issues in wireless sensor networks and many studies on power-saving in wireless sensor networks have been done. However, most of these studies have focused on saving power in gathering the information. In this paper, we tackle the issue of the energy-efficient information dissemination in wireless sensor networks. We propose method of disseminating information while optimizing electric power consumption in wireless sensor networks. Our new dissemination method employs flooding and uses the receiver signal strength to determine the broadcast timing. In our method, the node farthest from the sending node will rebroadcast the message first and a broadcast is canceled when a node scheduled to broadcast receives a duplicate of the scheduled message from other nodes. We evaluate our proposed method by simulation. As a result, it is found that the electric power consumption of the proposal method in the entire network is one-third that of the flooding method at the maximum. Furthermore, we show that the farther the transmitting distance of the wireless radio wave becomes, the more efficiently our method can disseminate information to the network.

A new available bandwidth measurement method based on ImTCP

We propose a method to measure available network bandwidth using the Internet Control Message Protocol (ICMP). The recently proposed ImTCP technique uses Transmission Control Protocol (TCP) data packets and the corresponding acknowledgement responses to measure the available bandwidth between sender and receiver. Since ImTCP needs to change the senders TCP implementation, it needs modifications to senders operating system kernel. Moreover, ImTCP cannot measure available bandwidth accurately if the receiver sends delayed acknowledgments. These problems stem from the use of TCP. In this paper, we discuss an ICMP-based method that overcomes these limitations. We evaluate the performance of the proposed method in an experimental network and show that it generates less measurement traffic and requires less time for bandwidth measurement than PathLoad. We also show that proposed method can measure the available bandwidth even if the bandwidth changes during measurement.

Congestion control scheme of Compound TCP+ in wireless LANs

Recently, Compound TCP has come to be regarded as the most promising transport layer protocol for high-speed and long-distance networks. In our previous work, we have shown that the loss-based congestion control mechanism performed by Compound TCP does not achieve fairness in throughput. In order to solve this problem, we have proposed Compound TCP+. This protocol decreases the loss window size when it is anticipated that a packet loss might occur, without an actual packet loss. Using simple simulations, we have shown that Compound TCP+ connections have high fairness in wireless LANs. However, in our previous work, we have not examined the congestion control scheme of Compound TCP+, and the performance of Compound TCP+ has been evaluated only by means of simple simulations. In this paper, we evaluate the congestion control scheme of Compound TCP+ in wireless LANs. We consider three forms of the congestion control scheme, and show that a linear decrease in the loss window size is good when the network is not in a state of serious congestion that causes a packet loss. Furthermore, we implement Compound TCP+ on a Linux kernel, and show its effectiveness by evaluating its performance in an experimental network environment.

Sizing Router Buffers for Large-Scale TCP/IP Networks

We investigate the validity of reducing router buffer size in a large-scale network that includes both core networks and edge networks. We first devise a novel mathematical analysis method of estimating the average behavior of TCP connections in a network with 100/1,000/10,000 routers/endhosts/links and 100,000 concurrent TCP connections. By applying our analysis method to the Abilene-inspired network, we demonstrate the influence of small buffer on link utilization, packet loss ratio and the performance of TCP connections passing through the router. One important result is that, especially when the edge network becomes faster, decreasing buffer size at core routers causes unfairness between TCP connections that traverse the core network and those do not traverse the core network.